Rare Presentation of an NF2-Mutated Desmoplastic Melanoma With Pure Spindle Cell Morphology in a 69-Year-Old Patient

To the Editor: Desmoplastic melanoma (DM) remains a challenging diagnosis with medico-legal pitfalls.1 Next-generation sequencing has become a critical tool in the bioinformatic interpretation of melanocytic neoplasms.2–6 Although pathogenic variants in NF1 are well-documented drivers in DM,7NF2 mutations are exceptionally rare in malignant cases and are more frequently associated with benign combined nevi.7 In this study, we present a novel case of a 69-year-old patient with an NF2-mutated desmoplastic melanoma exhibiting a pure spindle cell morphology. This case expands the recently described morphologic spectrum of NF-mutated neoplasms, challenging the prevailing genotype–phenotype correlations that associate NF2 truncations primarily with benign melanocytic neoplasms.7 Addo et al7 recently characterized the spectrum of NF-mutated melanocytic neoplasms and identified NF1 variants as the predominant driver in this cohort, present in most cases. In contrast, pathogenic variants in NF2 were identified in only 2 of 26 NF-mutated desmoplastic melanomas studied, with significantly more desmoplastic nevi harboring an NF2 mutation. In our presented case, we found a unique constellation not previously reported in the literature: A DM with NF2-mutation presenting with “pure” desmoplastic morphology and a melanoma in situ component, distinct from the benign NF2-mutated “clonal nevi” recently characterized in the literature. A 69-year-old female patient presented to a dermatologist in private practice with a slowly enlarging, nonpigmented indurated plaque on the left upper chest. The patient had no personal history of neurofibromatosis or malignant melanoma. A 3-mm punch biopsy was performed to rule out basal cell carcinoma. The initial 3-mm punch biopsy revealed mild dermal elastosis, and dermal spindle cells with scattered moderate cytologic enlargement and hyperchromasia, with a sparse lymphocytic infiltrate (Fig. 1). Immunohistochemically, the spindle cells were strongly positive for S100 and SOX10 but negative for CD34 and Melan-A. The following diagnosis was rendered: “neural tumor, superficial soft tissue tumor, and desmoplastic melanocytic proliferation cannot be ruled out, recommend re-excision for re-evaluation.”FIGURE 1.: H&E overview (left), H&E magnification (middle left), S100 stain (middle right), and high magnification SOX10 stain (right) of the initial punch biopsy. Scale bar, black, 100 μm.Re-excision demonstrated deeper dermal extension of the infiltrative lesion, falling short of subcutis. Hyperchromatic spindle cells were more frequent, without notable mitotic rate, and prominent perivascular lymphocytes, compared with the biopsy. Overlying epidermis showed increased concentration of slightly enlarged melanocytes compared with adjacent skin. PRAME immunohistochemistry revealed moderate-to-strong positivity in junctional melanocytes overlying the spindle cell proliferation, with an ill-defined demarcation toward the resection edges (Fig. 2).FIGURE 2.: SOX10 (left) and PRAME (right) staining of lesional spindle cells in the dermis and overlying melanocytes in the epidermis. Scale bar, black, 100 μm.Next-generation sequencing using the SureSelect Cancer CGP Assay8 was performed on the re-excision specimen. In this study, we found pathogenic mutations in NF2 (c.169C > T, truncating mutation), FGFR2 (c.1573G > T), and RAD51D (c.121C > T). Copy number variations were identified in PIK3CA (amplification, copy number 11.9), CDKN2A (loss, copy number 1.0), and MTAP (loss, copy number 1.0). There were VUS in CDC73, FLT1, and SNCAIP and a total TMB of 16.2 Mut/Mb. Taken together, a diagnosis of pure desmoplastic melanoma (pT3a) was rendered. The strongest supporting criteria were an overlying melanoma in situ component as well as prominent lymphocytic infiltrates, further supported by copy number variations including CDKN2A loss and PIK3CA amplification and a relatively high tumor mutational burden detected by next-generation sequencing. Wide local excision and sentinel lymph node dissection were performed following an institutional tumor board review. No evidence of residual tumor was identified in the margins or the lymph node. To sum up, this case represents an outlier from the recently established genotype–phenotype correlations for NF-mutated melanocytic tumors. According to Addo et al,7NF2 pathogenic variants are statistically more likely to be found in benign neoplasms than in melanomas. In their cohort of 30 NF-mutated cases, NF2 truncating variants were predominantly seen in a novel entity described as “NF-inactivated clonal nevus,” which typically features a combined epithelioid and spindle cell pattern without high-grade atypia. Two NF2-mutated desmoplastic melanomas were also described here: 1 harbored an additional NF1 mutation and the other was a pigmented, epithelioid desmoplastic melanoma mimicking a blue nevus. The identification of an NF2 truncating mutation in a high-grade, pure desmoplastic melanoma with no pigmentation and a melanoma in situ component underscores that this mutation is not exclusive to benign lesions and that more cases are needed to fully understand the spectrum of NF-mutated melanocytic lesions to establish the role of NF2 as a possible driver mutation in desmoplastic melanoma. This highlights the necessity of integrating genomic data with morphologic context and immunohistochemistry in large cohorts, as emphasized by recent literature, to avoid misclassifying unusual variants of desmoplastic neoplasms based on molecular data alone.